Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A method for executing computations in parallel for a building management system of a building, the method comprising: receiving, via a processing circuit of the building management system, a computing job request to determine values for one or more particular properties; receiving, via the processing circuit of the building management system, a property model indicating dependencies between a plurality of properties, the plurality of properties comprising the one or more particular properties, wherein the plurality of properties comprise raw properties; generating, via the processing circuit of the building management system, two or more computing threads based on the property model, wherein each computing thread comprises a sequence of computations for determining a value for each of the plurality of properties, wherein generating the two or more computing threads comprises: determining, for each property of at least some of the plurality of properties, a greatest dependency distance from the property to the raw properties; and ordering computations of the at least some of the plurality of properties in the two or more computing threads based on the greatest dependency distance of each of the at least some of the plurality of properties; and executing, via the processing circuit of the building management system, the two or more computing threads in parallel to determine the values for the one or more particular properties.
This invention relates to parallel computing techniques for building management systems, addressing the challenge of efficiently processing complex property dependencies in real-time. The system receives a computing job request to determine values for specific properties within a building management system. It also receives a property model that defines dependencies between multiple properties, including raw properties that do not depend on other properties. The system generates multiple computing threads, each containing a sequence of computations for determining property values. To optimize parallel execution, the system calculates the greatest dependency distance from each property to the raw properties, then orders computations within the threads based on these distances. This ensures that dependent properties are computed in the correct sequence while maximizing parallelism. The threads are then executed in parallel to efficiently determine the requested property values. The approach improves computational efficiency by leveraging parallel processing while respecting property dependencies, which is critical for real-time building management tasks such as energy optimization, HVAC control, and occupancy monitoring.
2. The method of claim 1 , further comprising controlling, via the processing circuit of the building management system, one or more pieces of building equipment of the building to control an environmental condition of the building based on the values for the one or more particular properties.
This invention relates to building management systems that control environmental conditions within a building. The system addresses the challenge of optimizing building operations by dynamically adjusting equipment based on real-time property values. A processing circuit within the building management system monitors and analyzes data from sensors or other sources to determine values for specific properties, such as temperature, humidity, or occupancy levels. The system then uses these values to control building equipment, such as HVAC units, lighting systems, or ventilation systems, to regulate environmental conditions like temperature, air quality, or lighting levels. The control logic ensures that the equipment operates in a coordinated manner to maintain desired conditions while improving energy efficiency and occupant comfort. The system may also incorporate predictive algorithms or historical data to anticipate changes in environmental conditions and preemptively adjust equipment settings. By integrating real-time data with automated control mechanisms, the invention enhances the responsiveness and efficiency of building management systems.
3. The method of claim 1 , wherein the dependencies between the plurality of properties comprise a computational relationship between one or more of the plurality of properties, wherein the computational relationship between the one or more of the plurality of properties indicates that a first value of a first property of the plurality of properties relies on a second value of a second property of the plurality of properties.
This invention relates to systems for managing and analyzing dependencies between properties in a computational environment. The problem addressed is the need to efficiently model and track relationships between multiple properties, where changes to one property may affect others. The solution involves defining computational relationships between properties, where the value of one property (the first property) depends on the value of another property (the second property). These relationships are used to ensure consistency and accuracy when updating or evaluating properties. The system may include a property management module that stores and retrieves property values, a dependency analyzer that identifies and resolves dependencies, and a computational engine that evaluates relationships to determine property values. The method involves detecting changes to a property, identifying dependent properties, and recalculating their values based on the defined relationships. This ensures that all properties remain consistent and up-to-date, improving reliability in systems where properties are interrelated. The approach is particularly useful in applications like configuration management, data modeling, and automated workflows where property dependencies are critical.
4. The method of claim 1 , wherein executing, via the processing circuit of the building management system, the two or more computing threads in parallel comprises performing one or more determinations of the value of each of the plurality of properties of the two or more computing threads and pausing execution of one of the two or more computing threads in response to determining that not all thread dependencies between the one of the two or more computing threads and another of the two or more computing threads for a current computation of the one of the two or more computing threads has been satisfied.
In the field of building management systems, efficient execution of computing threads is critical for optimizing performance and resource utilization. A method addresses the challenge of managing thread dependencies during parallel execution to prevent conflicts and ensure correct computation. The method involves executing multiple computing threads in parallel within a building management system, where each thread has properties that influence its execution. During parallel execution, the system evaluates the properties of each thread to determine whether all dependencies required for a current computation are satisfied. If a thread lacks necessary dependencies from other threads, its execution is paused until those dependencies are met. This ensures that computations proceed only when all required data or conditions are available, preventing errors and improving system reliability. The method dynamically monitors thread interactions, allowing for adaptive scheduling and conflict resolution in real-time. This approach enhances the efficiency and accuracy of parallel processing in building management systems, particularly in scenarios where threads rely on shared resources or sequential data.
5. The method of claim 1 , wherein the method further comprises: identifying, via the processing circuit of the building management system, recursion in the dependencies of the plurality of properties of the property model; and generating, via the processing circuit of the building management system, the two or more computing threads to include one or more computing steps associated with the recursion at an end of the two or more computing threads.
A building management system monitors and controls various properties of a building, such as temperature, lighting, and energy consumption. A challenge in such systems is efficiently processing complex dependencies between these properties, especially when recursive relationships exist, which can lead to computational inefficiencies or errors. This invention addresses this problem by improving the way the system processes property dependencies. The system uses a property model that defines relationships between multiple building properties. To handle recursion in these dependencies, the system identifies recursive loops within the property model. Once identified, the system generates multiple computing threads to execute the necessary steps for managing these properties. The steps associated with the recursion are strategically placed at the end of these threads to prevent infinite loops or redundant computations. This ensures that the system processes dependencies in a structured and efficient manner, avoiding conflicts or performance degradation. The approach enhances the reliability and responsiveness of the building management system by optimizing the handling of recursive property relationships.
6. The method of claim 5 , wherein the method further comprises: determining, via the processing circuit of the building management system, whether the two or more computing threads comprising the recursion create an infinite recursive loop; and returning, via the processing circuit of the building management system, an empty result for the plurality of properties associated with the infinite recursive loop.
This invention relates to building management systems that use recursive computing threads to process data. The problem addressed is the risk of infinite recursive loops in such systems, which can cause computational inefficiency, system crashes, or incorrect data processing. The invention provides a method to detect and handle infinite recursive loops in a building management system. The method involves using a processing circuit within the building management system to monitor the execution of two or more computing threads that perform recursive operations. The system determines whether these threads create an infinite recursive loop, which occurs when the recursion does not terminate as intended. If an infinite loop is detected, the system returns an empty result for the properties associated with that loop, preventing further processing errors. This ensures that the building management system remains stable and avoids incorrect data handling due to unchecked recursion. The method is part of a broader system that processes data related to building operations, such as energy management, HVAC control, or security monitoring. By detecting and mitigating infinite loops, the system maintains reliable performance and prevents disruptions in building automation tasks. The approach is particularly useful in environments where recursive algorithms are used to analyze complex data structures or dependencies within building systems.
7. The method of claim 1 , wherein the raw properties comprise data that has not been processed, wherein the raw properties comprise environmental data collected by building equipment of the building.
This invention relates to building management systems that process environmental data collected by building equipment to optimize operations. The problem addressed is the inefficient use of raw, unprocessed environmental data from building systems, which can lead to suboptimal energy consumption, comfort, and maintenance. The invention improves upon prior systems by leveraging unprocessed environmental data directly from building equipment, such as sensors and HVAC systems, to enhance decision-making. The method involves collecting raw environmental data from building equipment, which includes temperature, humidity, air quality, and other unprocessed measurements. This data is then analyzed to identify patterns, anomalies, or inefficiencies in building operations. The system may use this raw data to adjust equipment settings, predict maintenance needs, or optimize energy usage without requiring extensive preprocessing. By working with unprocessed data, the system avoids potential inaccuracies introduced by intermediate processing steps, leading to more precise and timely adjustments. The invention may also integrate with other building management functions, such as energy monitoring, fault detection, and predictive maintenance, to provide a comprehensive solution for improving building performance. The use of raw data ensures that the system responds dynamically to real-time conditions, enhancing efficiency and reducing operational costs.
8. The method of claim 7 , wherein executing, via the processing circuit of the building management system, the two or more computing threads in parallel comprises generating the value for each of the plurality of properties based on the dependencies between the plurality of properties; wherein the raw properties are first properties used in a first determination of the sequence of computations of the two or more computing threads, wherein the value for each of the plurality of properties depends on raw property values for the raw properties.
This invention relates to building management systems that optimize computational efficiency by executing multiple computing threads in parallel while managing property dependencies. The problem addressed is the need to efficiently determine the sequence of computations in a building management system where certain properties depend on others, requiring careful scheduling to avoid redundant or out-of-order calculations. The system includes a processing circuit that executes two or more computing threads in parallel to generate values for a plurality of properties. The properties include raw properties, which are used in an initial determination of the computation sequence. Each property's value depends on the raw property values of the raw properties, ensuring that computations are performed in an order that respects these dependencies. By processing threads in parallel while accounting for dependencies, the system avoids unnecessary recalculations and improves overall efficiency. The method ensures that property values are updated correctly, even when multiple threads are running simultaneously, by dynamically adjusting the computation sequence based on the latest raw property values. This approach optimizes resource usage and reduces latency in building management operations.
9. The method of claim 1 , wherein determining, via the processing circuit of the building management system, for each of the at least some of the plurality of properties, the greatest dependency distance from the property to the raw properties comprises: determining a number of dependencies between the property to each of the raw properties; and determining the greatest dependency distance from the property to the raw properties by comparing each of the number of dependencies between the property to each of the raw properties, wherein the greatest dependency distance is one of the number of dependencies that is a greatest number of dependencies.
In the field of building management systems, a method is disclosed for analyzing dependencies between properties and raw properties within a building automation system. The method addresses the challenge of efficiently determining the most significant relationships between derived properties and their underlying raw data sources. This is crucial for optimizing system performance, troubleshooting, and ensuring accurate data-driven decision-making in building operations. The method involves calculating the dependency distance between each property and the raw properties. For each property, the system determines the number of dependencies it has with each raw property. The greatest dependency distance is then identified by comparing these dependency counts, selecting the highest value as the greatest dependency distance. This approach quantifies how indirectly a property is derived from raw data, helping to map complex relationships within the system. The method supports improved system diagnostics, maintenance, and optimization by revealing the depth of data dependencies, ensuring that critical raw properties are properly monitored and maintained. This enhances the reliability and efficiency of building management operations.
10. A building management system of a building for executing computations in parallel, wherein the building management system comprises a processing circuit configured to: receive a computing job request to determine values for one or more particular properties; receive a property model indicating dependencies between a plurality of properties, the plurality of properties comprising the one or more particular properties, wherein the plurality of properties comprise raw properties; generate two or more computing threads based on the property model, wherein each computing thread comprises a sequence of computations for determining a value for each of the plurality of properties, wherein generating the two or more computing threads comprises: determining, for each property of at least some of the plurality of properties, a greatest dependency distance from the property to the raw properties; and ordering computations of the at least some of the plurality of properties in the two or more computing threads based on the greatest dependency distance of each of the at least some of the plurality of properties; and execute the two or more computing threads in parallel to determine the values for the one or more particular properties.
Building management systems (BMS) often require real-time computation of various properties (e.g., temperature, energy consumption) to optimize building operations. However, traditional systems struggle with efficiently processing interdependent properties, leading to delays or suboptimal performance. This invention addresses the problem by enabling parallel computation of property values based on their dependencies. The system includes a processing circuit that receives a computing job request to determine values for specific properties and a property model defining dependencies between multiple properties, including raw properties (e.g., sensor inputs). The circuit generates two or more computing threads, each containing a sequence of computations for determining property values. To optimize parallel execution, the system calculates the greatest dependency distance (the longest path from a property to its raw properties) for each property and orders computations in the threads based on this distance. This ensures that dependencies are resolved before computations proceed. The threads are then executed in parallel, allowing faster determination of the requested property values. By leveraging dependency-based threading, the system improves computational efficiency and reduces latency in building management operations. This approach is particularly useful for large-scale BMS where multiple interdependent properties must be computed simultaneously.
11. The system of claim 10 , wherein the processing circuit is configured to control one or more pieces of building equipment of the building to control an environmental condition of the building based on the values for the one or more particular properties.
This invention relates to building automation systems designed to optimize environmental control within buildings. The system addresses the challenge of efficiently managing building equipment to maintain desired environmental conditions, such as temperature, humidity, or air quality, while improving energy efficiency and occupant comfort. The system includes a processing circuit that monitors and analyzes data from sensors to determine values for specific environmental properties, such as temperature, humidity, or air quality. Based on these values, the processing circuit controls one or more pieces of building equipment, such as HVAC systems, ventilation units, or lighting, to adjust the environmental conditions of the building. The system dynamically adapts to changing conditions by continuously processing sensor data and adjusting equipment operation accordingly. This ensures that the building's environment remains within optimal parameters while minimizing energy consumption. The processing circuit may also incorporate predictive algorithms or machine learning to anticipate environmental changes and preemptively adjust equipment operation. Additionally, the system can integrate with user preferences or occupancy data to further refine control strategies. By automating environmental control, the system enhances energy efficiency, reduces operational costs, and improves occupant comfort.
12. The system of claim 10 , wherein the dependencies between the plurality of properties comprise a computational relationship between one or more of the plurality of properties, wherein the computational relationship between the one or more of the plurality of properties indicates that a first value of a first property of the plurality of properties relies on a second value of a second property of the plurality of properties.
This invention relates to a system for managing dependencies between properties in a computational environment. The system addresses the challenge of tracking and resolving interdependencies among multiple properties, where changes to one property may affect others. The system includes a dependency graph that maps relationships between properties, allowing for automated detection of conflicts or inconsistencies when values are updated. The dependencies are defined by computational relationships, meaning the value of one property (first property) depends on the value of another property (second property). For example, if a property A's value is derived from property B, the system ensures that any change to B is reflected in A. The system also includes a validation module to verify that property values comply with predefined rules, and a conflict resolution module to handle cases where dependencies lead to contradictory states. This approach improves system reliability by maintaining consistent property relationships and reducing manual intervention in dependency management. The system is particularly useful in software configurations, data modeling, and automated workflows where property interdependencies are complex and dynamic.
13. The system of claim 10 , wherein the raw properties comprise data that has not been processed, wherein the raw properties comprise environmental data collected by building equipment of the building.
This invention relates to a building management system that processes environmental data collected by building equipment to optimize building operations. The system addresses the challenge of efficiently managing building systems by leveraging raw, unprocessed environmental data from various building equipment, such as sensors, HVAC units, and other monitoring devices. The raw data includes temperature, humidity, air quality, and other environmental parameters that are directly collected without prior processing. The system processes this raw data to generate insights, detect anomalies, and make real-time adjustments to building operations, improving energy efficiency and occupant comfort. The system may also integrate with other building management components to provide a comprehensive solution for monitoring and controlling building environments. By utilizing unprocessed data, the system ensures accuracy and responsiveness in decision-making, reducing reliance on pre-processed or aggregated data that may lose critical details. The invention enhances building automation by enabling dynamic responses to environmental changes, optimizing resource usage, and maintaining optimal indoor conditions.
14. The system of claim 13 , wherein the processing circuit is configured to execute the two or more computing threads in parallel by generating the value for each of the plurality of properties based on the dependencies between the plurality of properties; wherein the raw properties are first properties used in a first determination of the sequence of computations of the two or more computing threads, wherein the value for each of the plurality of properties depend on raw property values for the raw properties.
This invention relates to a parallel computing system designed to optimize the execution of multiple computing threads by managing dependencies between properties. The system addresses the challenge of efficiently scheduling computations in parallel environments where properties have interdependencies, ensuring correct and timely execution without conflicts or redundant calculations. The system includes a processing circuit configured to execute two or more computing threads in parallel. The processing circuit generates values for a plurality of properties based on their dependencies. The system first identifies raw properties, which are properties used in an initial determination of the computation sequence for the threads. The values of these properties depend on the raw property values, meaning their computation must account for the dependencies to maintain accuracy and consistency. The processing circuit dynamically determines the sequence of computations for the threads, ensuring that dependent properties are computed in the correct order. This approach prevents race conditions and ensures that each thread accesses the most up-to-date property values. The system is particularly useful in high-performance computing, real-time data processing, and distributed systems where parallel execution must be both efficient and reliable. By managing dependencies explicitly, the system avoids unnecessary recomputations and minimizes latency, improving overall performance.
15. The system of claim 10 , wherein determining, for each of the at least some of the plurality of properties, the greatest dependency distance from the property to the raw properties comprises: determining a number of dependencies between the property to each of the raw properties; and determining the greatest dependency distance from the property to the raw properties by comparing each of the number of dependencies between the property to each of the raw properties, wherein the greatest dependency distance is one of the number of dependencies that is a greatest number of dependencies.
This invention relates to a system for analyzing dependencies between properties in a data processing environment, particularly for identifying the most significant relationships between derived properties and their underlying raw data sources. The system addresses the challenge of understanding how derived properties are constructed from raw data, which is critical for data lineage, debugging, and optimization in complex data pipelines. The system processes a plurality of properties, some of which are derived from raw properties, and determines the dependency distance between each derived property and the raw properties. Dependency distance measures the number of intermediate steps or transformations required to derive a property from raw data. For each derived property, the system calculates the number of dependencies to each raw property, then identifies the greatest dependency distance by comparing these values. The greatest dependency distance represents the longest path of transformations from a raw property to the derived property, helping to identify the most complex or indirect dependencies in the data pipeline. This analysis is useful for tasks such as impact analysis, where understanding the depth of dependencies helps assess how changes to raw data will propagate through the system. The system may also be used to optimize data processing by identifying redundant or overly complex transformations. The invention improves upon prior methods by providing a structured way to quantify and compare dependency distances, enabling better decision-making in data management.
16. A non-transitory computer-readable medium having machine instructions stored therein, the machine instructions being executable by a processor of a building management system to perform operations comprising: receiving a computing job request to determine values for one or more particular properties; receiving a property model indicating dependencies between a plurality of properties, the plurality of properties comprising the one or more particular properties, wherein the plurality of properties comprise building data for a building, wherein the plurality of properties comprise raw properties, wherein the raw properties comprise data that has not been processed, wherein the raw properties comprise environmental data collected by building equipment of the building; generating two or more computing threads based on the property model, wherein each computing thread comprises a sequence of computations for determining a value for each of the plurality of properties, wherein generating the two or more computing threads based on the property model comprises: determining, for each of the plurality of properties, a greatest dependency distance from the property to the raw properties; and ordering computations of the plurality of properties in the two or more computing threads based on the greatest dependency distance of each of the plurality of properties; and executing the two or more computing threads in parallel to determine the values for the one or more particular properties.
This invention relates to a building management system that processes building data efficiently by parallelizing computations based on property dependencies. The system addresses the challenge of managing large volumes of raw environmental data collected by building equipment, such as temperature, humidity, or energy consumption, which must be processed to derive meaningful insights. The raw data is unprocessed and requires complex computations to determine derived properties, such as energy efficiency metrics or comfort indices, which depend on multiple other properties. The system receives a request to compute values for specific properties and a property model that defines dependencies between various properties, including raw and derived properties. The property model is used to generate multiple computing threads, each containing a sequence of computations for determining property values. The system analyzes the dependencies to determine the greatest distance from each property to the raw data, then orders computations in the threads based on this distance to ensure dependencies are resolved before dependent properties are computed. These threads are executed in parallel, allowing the system to efficiently process the data and determine the requested property values. By leveraging parallel processing and dependency-based computation ordering, the system optimizes performance in building management applications where timely and accurate data processing is critical.
17. The non-transitory computer-readable medium of claim 16 , wherein executing the two or more computing threads in parallel comprises generating the values for the plurality of properties based on the dependencies between the plurality of properties; wherein the raw properties are first properties used in a first determination of the sequence of computations of the two or more computing threads, wherein the values for the plurality of properties depend on raw property values for the raw properties.
This invention relates to parallel computing systems that optimize the execution of multiple computing threads by dynamically determining the sequence of computations based on property dependencies. The problem addressed is the inefficiency in parallel processing when dependencies between properties are not properly managed, leading to redundant computations or delays in generating required values. The system involves a non-transitory computer-readable medium storing instructions that, when executed, enable parallel execution of two or more computing threads. The execution process generates values for a plurality of properties while considering the dependencies between them. The system first identifies raw properties, which are initial properties used to determine the initial sequence of computations for the parallel threads. The values of the plurality of properties are then derived based on the raw property values, ensuring that dependent properties are computed in the correct order to avoid conflicts or delays. By dynamically adjusting the computation sequence based on property dependencies, the system improves efficiency in parallel processing, reducing unnecessary computations and ensuring that dependent values are available when needed. This approach is particularly useful in applications requiring high-performance parallel execution, such as scientific simulations, data processing, or real-time systems.
18. The non-transitory computer-readable medium of claim 16 , wherein determining, for each of the at least some of the plurality of properties, the greatest dependency distance from the property to the raw properties comprises: determining a number of dependencies between the property to each of the raw properties; and determining the greatest dependency distance from the property to the raw properties by comparing each of the number of dependencies between the property to each of the raw properties, wherein the greatest dependency distance is one of the number of dependencies that is a greatest number of dependencies.
This invention relates to data processing systems that analyze dependencies between properties in a dataset. The problem addressed is efficiently determining the greatest dependency distance between derived properties and raw properties in a dataset, which is useful for understanding data relationships, feature selection, and model interpretability. The invention provides a method to quantify how indirectly a derived property depends on raw properties by counting the number of intermediate dependencies required to connect them. For each derived property, the system calculates the number of dependencies to each raw property, then identifies the maximum value among these counts as the greatest dependency distance. This helps identify which raw properties have the most indirect influence on derived properties, aiding in data analysis and model optimization. The approach is implemented in a computer-readable medium, ensuring reproducibility and scalability for large datasets. The solution improves upon prior methods by providing a clear, quantifiable metric for dependency depth, which is particularly valuable in machine learning and data science applications where understanding feature relationships is critical.
Unknown
April 7, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.